204 



National Resources Committee 



exclusively with agriculture, or labor, or commerce and 

 industry, or public health." 



Partly as a result of this lack of organized public 

 support and partly as a result of the frequent appoint- 

 ment of directors for political rather than professional 

 abilities, the Bureau of the Census has lacked money 

 and qualified personnel for keeping up with the mount- 

 ing needs of consumers of its products. 



Who Are the Consumers? 



The consumers may be divided into two broad 

 classes — public and private. The public consumers in- 

 clude most of the administrative arms of the Federal 

 Government, as well as State and municipal agencies. 

 The private consumers comprise university research 

 workers ; business and industrial concerns ; associations 

 like farm organizations, labor unions, and trade associa- 

 tions; teachers and journalists; and the miscellaneous 

 body of occasional users of statistics who may be 

 designated as the "general public." 



To both classes of consumers the importance of 

 census data has increased incalculably in the past half 

 century. It is true that during this period the share 

 of the Bureau of the Census in the pi'oduction of statis- 

 tics has declined, relative to the total volume of statis- 

 tics produced. But it is also true that the work of the 

 Bureau was never so important as it is today and that 

 the imjjortance of its functions is likely to increase 

 in the future. 



The Government as Consumer 



Let us look backward to the 1880's, when Francis 

 Walker's second census was taken. 



The Federal Government had little need for statis- 

 tics in carrying out its relatively simple functions. The 

 service to agriculture, for example, was in its infancy, 

 and there was no program of crop control and sub- 

 sidies, calling for accurate measures of acreage and 

 production. Information about acreage and produc- 

 tion was interesting even then, in order to give a 

 "picture" of the rural economy. Today it is not only 

 interesting; it is indispensable to the execution of legis- 

 lation involving hundreds of millions of dollars. Con- 



» "The usual test of a business enterprise," said Dr. Willard C. 

 Thorp, of Dun & Bradstreet, recently, "is market absorption of its 

 product. The number of satisfied customers is indicated by its sales 

 volume. Statistical work by Government agencies Is In part for pur- 

 poses of determining Government policy, but much of it must be justified 

 also on the basis of Its usefulness to outsiders. These consumers do 

 not record themselves as they would by purchasing commodities • • • 

 We use census data but never report their value to the Census Bureau 

 for support In Its annual budget battle • • • Demonstrations by 

 the consumers of statistics are of tremendous importance to those re- 

 sponsible for planning the program and the expenditures of Government 

 agencies. That support is essential Is already Indicated by the mis- 

 fortune of the CSB (Central Statistical Board) in the sharp curtailment 

 of Its budget by Congress." In a Review In Journal of American Statis- 

 tical Association, XXIII (March 1938), 270. 



sequently, the demands for detail and accuracy are tre- 

 mendously greater today. 



Again, consider the responsibilities assumed by the 

 Government in the 1880's with respect to regulation of 

 industry, trade, and labor conditions. Except for the 

 tariff, the policy was mainly one of laissez-faire. 

 Today, the Government has undreamed-of regulatory 

 responsibilities. Accurate data on manufactures, 

 transportation, power, trade, wages and hours of labor, 

 and the industrial, occupational, and employment 

 classification of the population are not merely, as in 

 Walker's day, desired primarily to i^resent a "pic- 

 ture" of the industrial economy. The data are indis- 

 pensable to Government. While agencies other than 

 the Bureau of the Census compile most of the cur- 

 rent (annual, monthlj') industrial and labor statis- 

 tics, the census of manufactures, together with the 

 censuses of population, agriculture, and distribution, 

 provides basing points without which satisfactory 

 current series could not be constructed. 



Or, we may note that in the 1880's the idea of Social 

 Security as a function of the Federal Government had 

 hardly occurred. The census of population was of 

 great local interest to fast-growing cities boastful of 

 their progress and was of value to writers who sought 

 to describe the development of the Republic. But, 

 apart from its Constitutional function of determining 

 Congressional apportionment, the census of population 

 was probably more of a luxury than a necessity to the 

 Federal Government itself. The census was a social 

 service rendered by the Government. Gradually, its 

 usefulness to Government increased, but the curve of 

 its importance made its greatest upward leap with the 

 passage of the Social Security Act only four years 

 ago. Almost overnight the census of population be- 

 came basic to policies involving the welfare of millions 

 of citizens and involving the ultimate handling of bil- 

 lions of dollars. At once, the need of the Govern- 

 ment for a quinquennial rather than a decennial census 

 became imperative. At the same time, inadequacies of 

 past censuses, which were not designed, of course, with 

 a Social Security Act in mind, became apparent. Ac- 

 tuarial work in the Social Security Board depends on 

 knowledge of the age and sex distribution of the pop- 

 ulation, by industrial, occupational, and employment 

 status. Administrative work depends on knowledge of 

 the regional distribution of employers, classified by 

 number of employees, on knowledge of family compo- 

 sition, and on distinctions in the census between "cov- 

 ered" and "noncovered" industries, while Federal 

 grants to States require elaborate analyses of census 

 figures. Finally, proposed changes in the law will call 

 for research on their probable cost and consequences, 

 and this research will rest upon the base of census data. 



